I went 446/1664 or 26.80%. Anything above 25% is nice, but this definitely wasn’t what I was expecting from a 3x event. The confidence interval (95%) with my sample size is 24.67% to 28.93%, so the expected 30% isn’t even within range. So, fairly unlucky. Bleach skill ups, combined with the horrible drop rates, were particularly bad so I’ll be using Pys on the rest of those.

### Like this:

Like Loading...

Nowhere near as many samples, but this event was massively disappointing bordering on enraging for me: 29/128, or 22.66%, so not even as good as 2.5x.

What calculator are you using? http://statpages.info/confint.html has your 99% confidence interval at 24.05% to 29.69% and http://stattrek.com/online-calculator/binomial.aspx has the probability of getting 446 or fewer successes in 1664 trials as 1 in 451. I’d say that’s a lot more than “fairly” unlucky.

LikeLike

I was using a 95% confidence interval, but yeah, I saw what it was at 99%. I tend to downplay luck in either direction since variance spikes like this happen from time to time, but I admit it’s unlikely that I’ll dig myself out of this hole for a long time.

LikeLike

Are you always feeding 5 at once? And if so, are you accounting for overfeeding? If so, it’s possible that the rate is still 30%, but some of those skillups are occurring when a monster is max skilled already.

LikeLike

There would be no point in recording these numbers if I didn’t factor that in.

LikeLike

Sorry, I should have asked “how are you accounting for overfeeding?” I ask because I couldn’t think of a good way of doing this correctly without being either

a) very time-intensive – you make sure you never overfeed by feeding at most x monsters when you have x skillups left.

b) throwing away all data whenever you fuse and have the chance of overfeeding. The main thing I’m worried about here is if you, say, feed 5 to something with 3 skillups left and you try to adjust your numbers after the fact. Ex: Only count the feed if if you get at most 2/5.

LikeLike

Pretty sure he just said that he accounted for that.

i.e. If there is one skill up left, he would feed one at a time or else the data he provided will not be accurate.

LikeLike

(a) is the only way to do it accurately. (b) is also fine, but would lower sample size. Time is indeed a factor and is the reason why I don’t record this data anymore, I just did it on a whim since it was the first 3x event.

LikeLike

Waiting on the next event for EXP bonus multipliers to spend all my king dragons. Do you know how likely getting 1.5x and 2x bonuses are usually? I’ve hunted around for that info a bit and haven’t found anything.

LikeLike

Sorry, I don’t.

LikeLike

For any more skillups would you wait for the next 3x, or just go for the next 2.5x? I valued 10x more than 3x skillups, so now I’m slowly filling my box up again 😦

LikeLike

It’s hard to say for sure since there’s no telling when we’ll get it again. For me, 2.5x vs. 3x doesn’t really matter. I can see it paying off for other players, though, especially if box space isn’t a huge issue.

LikeLike

What difficulty were you doing for the 6* bleach cards (Ichigo & Rukia) skill ups? I haven’t gotten many to drop…

LikeLike

I did 50 stam for the first Rukia, then just Py the rest.

LikeLiked by 1 person

Do you think they changed the rate from 10% to 9%? There have also been other large posts on Reddit and such that got a lower rate, although there’s some sampling bias involved here

LikeLiked by 1 person

I don’t think so.

LikeLike